WarDragon Updates for DragonOS R36 Released…<\/a><\/p>\n\n\n\n Feature Review: Tempest…<\/a><\/p>\n\n\n\n Explore More with DragonOS…<\/a><\/p>\n\n\n \n\t\t\t\t\t\tSubscribe<\/a>\n\t\t\t\t\t<\/p>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n\n\n\n\n Canine Defense Technologies<\/a> aims to \u201cbridge the U.S. Defense Tech hardware gap.\u201d This week they released the video below showing the Dragon Drone in flight: https:\/\/x.com\/K9DefenseTech\/status\/1815904436469793026<\/a><\/p>\n<\/div>\n\n\n\n The Dragon Drone is a project that is still in development through a collaboration between Cemaxecuter, Canine Defense Technologies, and ARK Electronics<\/a>. The objective of the project is to increase radio survey capabilities within DragonOS through a highly mobile platform. While the WarDragon can be carried or driven around, the DragonDrone can quickly deploy from the ground to high altitudes for favorable RF reception and traverse space in a more flexible manner than the WarDragon.<\/p>\n ARK Electronics provides the flight computer through its Jetson Orin PAB Carrier<\/a>, which also enables a Jetson Orin running DragonOS to operate on board. Canine Defense is doing a great job integrating the aircraft body and electronics – we\u2019re looking forward to more updates including specifications, hardware detail shots, and more videos as the drone continues towards its release. The best place to monitor updates is by following Cemaxecuter<\/a> and Canine Defense Technologies<\/a> on X.<\/p>\n<\/div>\n\n\n\n Last week<\/a>, we mentioned that the WarDragon is due for its update to DragonOS Focal R36. That update is now released. Don\u2019t worry, this one doesn\u2019t create a kernel panic, let alone a global one, like CrowdStrike\u2019s update that went out to Windows systems this past week. If you own a WarDragon and did not receive update instructions from Cemaxecuter, please reach out to him directly via the contact information below<\/a> to resolve your issue.<\/p>\n Included with the update is a sha256sum hash for the update file as well as a video for how to update the computer inside the WarDragon. After downloading both the download file and hash file, you can verify the integrity of the download by running the command below:<\/p>\n The output should appear as follows:<\/p>\n If you see the below output, please contact Cemaxecuter directly because the download file\u2019s hash does not match the original value:<\/p>\n When updating your WarDragon, you will not<\/strong> use the USB hub to conduct the reimage. The video instructions for updating your WarDragon will alert you to this. The drive and Clonezilla Live USB will need to be connected directly to the WarDragon\u2019s USB ports.<\/p>\n The process will use Clonezilla to reimage the WarDragon to R36<\/a>, so before you update, back up any key files you wish to take to the new update. DragonOS comes with Timeshift<\/a> by default, so you can use that tool if you prefer.<\/p>\n<\/div>\n\n\n\n If you prefer, you can send Cemaxecuter a drive with return shipping for a free upgrade as shown in the X post below: https:\/\/twitter.com\/cemaxecuter\/status\/1813364075914547403<\/a><\/p>\n<\/div>\n\n\n\n Wikipedia defines TEMPEST<\/a> as a \u201ca U.S. National Security Agency specification and a NATO certification referring to spying on information systems through leaking emanations, including unintentional radio or electrical signals, sounds, and vibrations.\u201d The specification also includes how to prevent equipment from spying attempts. In 2014, Martin Marinov developed TempestSDR<\/a> which brought the capability of eavesdropping on HDMI, DVI, and VGA to SDRs such as the SDRPlay and Ettus series. TempestSDR ships with DragonOS Focal R36.<\/p>\n In 2021, Federico Larroca implemented TempestSDR in GNU Radio, creating gr-tempest<\/a>. gr-tempest also ships with DragonOS Focal R36 and you can watch Cemaxecuter demonstrate the capability of eavesdropping with gr-tempest on electronic equipment emissions with a RTL-SDR<\/a> or\neavesdropping on a monitor with SDRPlay<\/a>. This out-of-tree (OOT) module was updated within DragonOS to version 3.10.10 a couple of weeks ago<\/a> with the rest of GNU Radio\u2019s OOT modules. gr-tempest essentially makes the usage of Martin\u2019s original TempestSDR in GNU Radio, which is a nice feature to have for those that are used to the GNU Radio development ecosystem. If you are interested in how gr-tempest works, you can watch Federico deliver a talk<\/a> during GNU Radio\u2019s 2021 Conference about his project.<\/p>\n In a separate project in 2024, Martin\u2019s TempestSDR also became the basis of EMEye<\/a>. EMEye was developed by Yan Long and Qinhong Jiang, students at the University of Michigan and Zhejiang University, respectively. Their work extended the capability of EMEye to detect images from specific camera models that have spurious radio frequency emissions. DragonOS Focal R36 ships with EMEye and you can watch Cemaxecuter demonstrate this experimental research against one of his vulnerable cameras<\/a>.<\/p>\n Recently, researchers from the Facultad de Ingenieria (a.k.a Fing) in Uruguay developed Deep tempest<\/a>. Their draft paper can be found on Arxiv<\/a> and their experiments show Deep Tempest can reduce the character error rate for displaying words from digital video images using only spurious RF emissions. Deep Tempest does not yet ship with DragonOS, but you can experiment with the researchers code base using their GitHub repository: https:\/\/github.com\/emidan19\/deep-tempest<\/a>. Deep Tempest improves on TempestSDR by using the DRUNet convolutional neural network (CNN) to infer the original video image from a set of training data rather than relying on AM demodulation and channel equalization. Deep Tempest also claims to make quality-of-life improvements on gr-tempest and TempestSDR by automatcially determining the best center frequency for image decoding and automatically determining the start of image frames by avoiding the blanking period that digital video signals use for audio and control data.<\/p>\n Deep Tempest relies on gr-tempest to train the CNN and for the time-being is only meant to infer text-based images. Therefore, it is suggested that users employ a GPU to accelerate the training time for Deep Tempest. Furthermore, while the developers included sample training data for experimentation, this training data only pertains to the image settings that the developers used for their tests: 1920x1080p at 60 Hz. Any modifications to these settings would likely throw off the inference accuracy. Lastly, even after users obtain training data, Deep Tempest is not able to infer images in real-time.<\/p>\n<\/div>\n\n\n\n
\n\n\n\nHighlight of the Week: Dragon Drone Takes Flight<\/h2>\n\n\n\n
![\"\"](\"https:\/\/cemaxecuter.com\/wp-content\/uploads\/2024\/07\/canine.png\")
<\/figure>\n\n\n\n
\n\n\n\nWarDragon Updates for DragonOS R36 Released<\/strong><\/h2>\n\n\n\n
sha256sum --check WarDragon_2024-07-16-11-img.tar.gz.sha256.txt\n<\/code><\/pre>\ndragon@dragon:~\/Downloads$ sha256sum --check WarDragon_2024-07-16-11-img.tar.gz.sha256.txt \nWarDragon_2024-07-16-11-img.tar.gz: OK\n<\/code><\/pre>\ndragon@dragon:~\/Downloads$ sha256sum --check WarDragon_2024-07-16-11-img.tar.gz.sha256.txt \nWarDragon_2024-07-16-11-img.tar.gz: FAILED\nsha256sum: WARNING: 1 computed checksum did NOT match\n<\/code><\/pre>\n![\"\"](\"https:\/\/cemaxecuter.com\/wp-content\/uploads\/2024\/07\/cemaxecuter-1.png\")
<\/figure>\n\n\n\n
\n\n\n\nFeature Review: Tempest<\/h2>\n\n\n\n
![\"\"](\"https:\/\/cemaxecuter.com\/wp-content\/uploads\/2024\/07\/tempest-comparison.png\")
\n\n\n\nExplore More with DragonOS<\/h2>\n\n\n\n
\n
\n
\n
\n
\n
\n
\n
\n\n\n